SF20K-Test
Models answer questions about the story in a short film. The authors check multiple-choice answers against a key and grade free responses with GPT-3.5.
The authors compare 13 models. The test contains 4,885 questions about 1,072 films.
- Made by
- École Polytechnique and MBZUAI
- Released
- June 2024
- Kind
- Long videos
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision
- Last model added
- September 2024
- Scores refreshed
- November 20th, 2026
GPT-4o-mini (snapshot unspecified) tops the leaderboard with 69.7% LongVA-7B leads with 62.8% among models released by September 2024
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. The headline averages multiple-choice and open-ended accuracy for the same input setting. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is LongVA-7B, at 62.8%. This view shows only models we track, each at its best.
The newest model with a score here came out in September 2024, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1 GPT-4o-mini (snapshot unspecified) Vision-language 69.7%
- 2 LongVA-7B Language-only 62.8%
- 3
Pixtral 12B Vision-language 62.1%
- 4 Llama-3.2-Vision 11B Vision-language 61.4%
- 5
LLaVA-OneVision 7B Vision-language 58%
- 6 long-llava-qwen2-7b Language-only 53.5%
- 7 Llava-Next-Video 7B Vision-language 53.5%
- 8
mPLUG-Owl2 Vision-only 30.2%
- 9
LLaVA-OneVision 0.5B Vision-language 30.2%
- 10 Video-LLaVA-7B Vision-language 26.9%
- 11 TimeChat 7B Vision-only 25.9%
- 12 MovieChat 7B Vision-only 11.2%
Source: Official SF20K-Test leaderboard
- 1 LongVA-7B Language-only 62.8%
- 2
Pixtral 12B Vision-language 62.1%
- 3
LLaVA-OneVision 7B Vision-language 58%
- 4 long-llava-qwen2-7b Language-only 53.5%
- 5
mPLUG-Owl2 Vision-only 30.2%
- 6
LLaVA-OneVision 0.5B Vision-language 30.2%
- 7 Video-LLaVA-7B Vision-language 26.9%
Source: Official SF20K-Test leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-4o-mini (snapshot unspecified) | 69.7% | Vision-language | Official leaderboard | |
| 2 | LongVA-7B | 62.8% | Language-only | Jun 2024 | Official leaderboard |
| 3 | | 62.1% | Vision-language | Sep 2024 | Official leaderboard |
| 4 | Llama-3.2-Vision 11B 11B | 61.4% | Vision-language | Official leaderboard | |
| 5 | | 58% | Vision-language | Aug 2024 | Official leaderboard |
| 6 | long-llava-qwen2-7b | 53.5% | Language-only | Aug 2024 | Official leaderboard |
| 7 | Llava-Next-Video 7B 7B | 53.5% | Vision-language | Official leaderboard | |
| 8 | | 30.2% | Vision-only | Nov 2023 | Official leaderboard |
| 9 | | 30.2% | Vision-language | Aug 2024 | Official leaderboard |
| 10 | Video-LLaVA-7B | 26.9% | Vision-language | Nov 2023 | Official leaderboard |
| 11 | TimeChat 7B 7B | 25.9% | Vision-only | Official leaderboard | |
| 12 | MovieChat 7B 7B | 11.2% | Vision-only | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | LongVA-7B | 62.8% | Language-only | Jun 2024 | Official leaderboard |
| 2 | | 62.1% | Vision-language | Sep 2024 | Official leaderboard |
| 3 | | 58% | Vision-language | Aug 2024 | Official leaderboard |
| 4 | long-llava-qwen2-7b | 53.5% | Language-only | Aug 2024 | Official leaderboard |
| 5 | | 30.2% | Vision-only | Nov 2023 | Official leaderboard |
| 6 | | 30.2% | Vision-language | Aug 2024 | Official leaderboard |
| 7 | Video-LLaVA-7B | 26.9% | Vision-language | Nov 2023 | Official leaderboard |
Rankings by part
We read 2 parts from the benchmark’s own leaderboard: Multiple-choice question answering and Open-ended question answering. Each ranks models on one part of the test. LLaVA-OneVision 7B leads Multiple-choice question answering and LongVA-7B leads Open-ended question answering. Arrows show how far a model moved from its overall rank among the models we track.
Multiple-choice question answering
7 models we track have a score here.
- 1
LLaVA-OneVision 7B ↑2 78.2%
- 2
Pixtral 12B 74.4%
- 3 LongVA-7B ↓2 73.8%
- 4 long-llava-qwen2-7b 72.6%
- 5 Video-LLaVA-7B ↑2 45.7%
- 6
mPLUG-Owl2 ↓1 38.3%
- 7
LLaVA-OneVision 0.5B ↓1 37.8%
Open-ended question answering
7 models we track have a score here.
- 1 LongVA-7B 51.7%
- 2
Pixtral 12B 49.8%
- 3
LLaVA-OneVision 7B 37.7%
- 4 long-llava-qwen2-7b 36.7%
- 5
LLaVA-OneVision 0.5B ↑1 22.6%
- 6
mPLUG-Owl2 ↓1 22.1%
- 7 Video-LLaVA-7B 10.6%
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: Official SF20K-Test leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| LongVA-7B | Language-only | 62.8% | |||
| Vision-language | 62.1% | ||||
| Vision-language | 58% | ||||
| Language-only | 55% | ||||
| long-llava-qwen2-7b | Language-only | 53.5% | |||
| LongVA-7B | Vision-language | 53.4% | |||
| long-llava-qwen2-7b | Vision-language | 53% | |||
| Vision-only | 44.1% | ||||
| LongVA-7B | Vision-only | 42.4% | |||
| long-llava-qwen2-7b | Vision-only | 34.8% | |||
| Vision-only | 34% | ||||
| Vision-only | 30.2% | ||||
| Vision-language | 30.2% | ||||
| Language-only | 27.8% | ||||
| Video-LLaVA-7B | Vision-language | 26.9% | |||
| Vision-only | 21.8% | ||||
| Video-LLaVA-7B | Vision-only | 20.4% | |||
| Video-LLaVA-7B | Language-only | 15.9% | |||
| Vision-language | 11.4% | ||||
| Language-only | 11.3% |
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: Official SF20K-Test leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| LongVA-7B | Language-only | 62.8% | |||
| Vision-language | 62.1% | ||||
| Vision-language | 58% | ||||
| Language-only | 55% | ||||
| long-llava-qwen2-7b | Language-only | 53.5% | |||
| LongVA-7B | Vision-language | 53.4% | |||
| long-llava-qwen2-7b | Vision-language | 53% | |||
| Vision-only | 44.1% | ||||
| LongVA-7B | Vision-only | 42.4% | |||
| long-llava-qwen2-7b | Vision-only | 34.8% | |||
| Vision-only | 34% | ||||
| Vision-only | 30.2% | ||||
| Vision-language | 30.2% | ||||
| Language-only | 27.8% | ||||
| Video-LLaVA-7B | Vision-language | 26.9% | |||
| Vision-only | 21.8% | ||||
| Video-LLaVA-7B | Vision-only | 20.4% | |||
| Video-LLaVA-7B | Language-only | 15.9% | |||
| Vision-language | 11.4% | ||||
| Language-only | 11.3% |
2 models have held number one on the official leaderboard since its release in June 2024
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official SF20K-Test leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official SF20K-Test leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Audio and video
- Video-MMMU In use 300 expert-level lecture videos with questions testing perception, comprehension and adaptation of knowledge. Accuracy.
Qwen 3.8 Max 88.7%
- LVBench In use Hours-long video understanding with multiple-choice questions on reasoning, grounding and summarization.
Gemini 4 Argon 91.7%
- MMVU In use 3,000 expert-annotated questions on videos across 27 disciplines. Multiple-choice and open-ended accuracy.
Qwen 3.8 Max 82.4%
- MLVU In use Multi-task long video understanding (minutes to hours) with multiple-choice and generation tasks. Accuracy.
Qwen 3.8 Max 90.8%
- FLEURS In use Models transcribe read speech in 102 languages built from FLoRes-101 sentences. Model cards usually report word error rate, so lower scores are better.
Phi-4-multimodal 4.00%
- VoiceBench In use Spoken instructions with speaker, environment and content variation. Voice assistant responses scored by GPT judge or rules.
Qwen3.5-Omni-Plus 93.1%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 2 years ago, in September 2024, for Pixtral 12B. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the official leaderboard (12 entries).
- Pass Models scored 7 models we track, plus 5 other entries on its leaderboard.
- Warning Newest score The newest came 2 years ago, in September 2024, for Pixtral 12B.
- Note Frontier models None of its models are on the capabilities index.
- Pass Headroom The best is 62.8%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on ridouaneg.github.io: 20 rows for models we track and 5 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.